Minutes for LMAP at interim-2014-lmap-2
Large-Scale Measurement of Broadband Performance
||Minutes for LMAP at interim-2014-lmap-2
LMAP WG Interim Meeting
Date and Time: Monday December 15, 2014, noon EST
Chairs: Dan Romascanu, Jason Weil
Based on notes taken by Barbara Stark
1. Note Well, Note Takers, Jabber Scribes, Agenda Bashing - Chairs (5 min)
2. WG Status - Chairs (5 min)
3. Use Case document in IESG review: input from Benoit and discussions (30 min)
4. LMAP Framework - edit proposals to solve AD comments: Al Morton (20 min) 5.
LMAP Information Model - Trevor, Phil: discuss proposals presented at IETF-91
(30 min) 6. Protocol Selection Criteria - Barbara, Tim (30 min) 7. Next steps
Use Case Document:
Look at Discuss comments.
Discussion of Pete Resnick Discuss (msg01925).
Action Item: Frode is to rewrite the regulatory use case to reverse the order
in which aspects of the use case are introduced and reduce the amount of
political motivation mentioned. Pass it by Pete before bringing it to the wider
Discussion of Kathleen Moriarty's Discuss (msg01926)
Phil Eardley described his resolution of Kathleen's points.
Kathleen needs to see resolutions and state whether they are acceptable.
LMAP Framework Document:
Phil described changes to framework draft from -08 to -09 version to address
Benoit's AD comments, Radia's security review comments, and comments from
Dublin discussion. Aamer Akhter will be providing some review. Need to identify
what's missing at this time. It's unclear. Need to talk further with Benoit to
understand. Benoit to organize a meeting with himself, chairs, and framework
editors. Anyone else who wants to contribute will be welcome. Benoit will find
a time with chairs and editors and then invite all.
Phil was not prepared to discuss information-model
Tim Carey presented slides on criteria for protocol selection.
Change "ensure interoperability" to "test interoperability"
Tim and Barbara to create IETF draft of these criteria. This will allow for
proper WG review. Alissa had comments regarding meaning of requirement
regarding security. Per Tim, encryption and authentication through use of
information-model certificates. What about confidentiality? Read RFC 3552.
Marcelo mentioned there is a difference in supporting certificates from MA vs.
Controller side. Complexity of MA certificates may not be mandatory, but could
differentiate. It may be enough to say the protocols need to support the
security features described in the framework document, as the mandatory aspect,
and allow other aspects to differentiate. Need to understand which aspects of
security are mandatory and which are optional.
Next Interim: Dan will set up Doodle poll to schedule another meeting on a day
during the first 2 weeks of February.